# Evidence Library: Index of Test Reports R1–R9

> The source of every key figure on this site: scope, platform and findings of nine test reports across AMD MI308X, Huawei Ascend 910B and MetaX N260, including a third-party report.

URL: https://vuke.top/en/evidence · Updated November 12, 2026

Every key figure on this site is followed by an ID that points to a report listed here. This page publishes what each report tested, on which platform, and what it concluded; the original documents are available on request.

## Test reports

| ID | Date | Type | Test platform | Title |
| --- | --- | --- | --- | --- |
| [R1](https://vuke.top/en/evidence/R1) | 2026-07-03 | Benchmark report | 8 × AMD Instinct MI308X · ROCm 7.2 | FX100 Comprehensive LLM Inference and Training Benchmark (8 × AMD MI308X) |
| [R2](https://vuke.top/en/evidence/R2) | 2026-07-05 | Benchmark report (official edition) | 8 × AMD Instinct MI308X · ROCm 7.2 | FX100 KV Cache Benchmark (480B, TP8, Long Context) |
| [R3](https://vuke.top/en/evidence/R3) | 2026-07-06 | Benchmark summary report (official edition) | 8 × AMD Instinct MI308X · ROCm 7.2 | FX100 KV Cache Benchmark Summary (480B, TP4×2, All Metrics) |
| [R4](https://vuke.top/en/evidence/R4) | 2026-07-06 | Official third-party test report | 8 × AMD Instinct MI308X · ROCm 7.2 | FX100 KV Cache Benchmark (480B, Multi-Instance, No. -006) |
| [R5](https://vuke.top/en/evidence/R5) | 2026-07-03 | Benchmark report (official edition, No. -004) | Single MetaX N260 GPU | FX100 KV Cache Benchmark (14B, HBM Efficiency) |
| [R6](https://vuke.top/en/evidence/R6) | 2026-07-07 | Deployment and adaptation report | AMD Instinct MI308X · ROCm 7.2 | ComfyUI + LTX-Video 2.3 Full-Model Deployment, Adaptation and Usage Report (V2) |
| [R7](https://vuke.top/en/evidence/R7) | 2026-07-07 | Model adaptation report | AMD Instinct MI308X | ComfyUI + LTX-Video 2.3 Model Adaptation Report (AMD MI308X) |
| [R8](https://vuke.top/en/evidence/R8) | 2026-07 | Code export and raw data | AMD Instinct MI308X · ROCm 7.2 | FX100 KV Cache Test Code Export and Raw Benchmark Data |
| [R9](https://vuke.top/en/evidence/R9) | 2026-05-30 | Benchmark report | 8 × Huawei Atlas 910B (Kunpeng-920) | FX100 on Huawei Ascend 910B: Inference and Training Benchmark (vs. NFS Baseline) |

## Key figures and their sources

| Metric | Value | Baseline | Detail | Source |
| --- | --- | --- | --- | --- |
| Higher inference throughput | +29–40% | vs. a single local NVMe drive | Qwen3-Coder-480B-FP8 in TP=8 production topology under a long-context cold-recovery load: +29% at concurrency 8 and +40% at concurrency 16, the operating point with the largest gain; +35–36% for the whole machine in the dual-instance TP4×2 layout. | Measured · [R2](https://vuke.top/en/evidence/R2) [R3](https://vuke.top/en/evidence/R3) |
| Lower time to first token (TTFT) | −26–32% | vs. a single local NVMe drive | Across three concurrency levels of the same 480B TP8 load, TTFT p50 falls from 10.17–35.73 s to 7.53–26.35 s. | Measured · [R2](https://vuke.top/en/evidence/R2) |
| Faster than recomputing without external storage | 8.6–20× | vs. full recompute with no external storage | At concurrency 16, TTFT p50 is 149.5 s when recomputing and 11.85 s with FX100; throughput is 4.1 tok/s versus 74.9 tok/s. | Measured · [R2](https://vuke.top/en/evidence/R2) |
| TTFT gain from the LMCache parallel-read patch | 4.1× | vs. before the patch | One GPU, concurrency 16, cold read from disk (Qwen2.5-32B): TTFT 37.97 s → 9.30 s; read bandwidth 0.98 → 5.23 GB/s, about 5.3×. | Measured · [R1](https://vuke.top/en/evidence/R1) |
| Faster model loading | 6.2–9.3× | vs. NFS, on Huawei Ascend 910B | On Huawei Atlas 910B: DeepSeek-32B serving load 691 s → 112 s (6.2×); DeepSeek-70B 1399 s → 150 s (9.3×). | Measured · [R9](https://vuke.top/en/evidence/R9) |
| Faster training checkpoint saves | 1.9× | vs. the baseline in report R1 | 8-GPU 32B LoRA with 65.6 GB full-model snapshots: 178 s → 94 s; sustained write bandwidth 3.26 → 6.40 GB/s (+96%). | Measured · [R1](https://vuke.top/en/evidence/R1) |
| Effective compute utilization under model switching | 46.7% → 62.8% | at 20 model switches per hour | Modeled from measured load times: at 20 model switches per hour, effective compute utilization rises from 46.7% to 62.8%. | Measured · [R1](https://vuke.top/en/evidence/R1) |
| Concurrent model loading on 8 GPUs | 16% / 30% | vs. the baseline in report R1 | Eight GPUs cold-reading models at once: 16% faster wall-clock overall and 30% faster per model. | Measured · [R1](https://vuke.top/en/evidence/R1) |
| Deviation between two independent runs | ≤5% | same load level, measured twice | Two independent runs at the same load level deviate by no more than 5%; cross-instance hot sharing through the fs:// shared pool was verified in the same campaign. | Measured · [R3](https://vuke.top/en/evidence/R3) |
| HBM residency an external KV tier can stand in for | 128 GB | MetaX N260 single GPU, 14B model, seven-way comparison | Conclusion of a seven-way comparison on a 14B model: an external KV tier can stand in for 128 GB-class HBM residency. That campaign ran on a single MetaX N260 GPU; the method carries across platforms, the numbers are not mixed with MI308X results. | Measured · [R5](https://vuke.top/en/evidence/R5) |
| ComfyUI + LTX-Video 2.3 models running end to end | 7 / 7 | AMD MI308X, ROCm 7.2 | All 7 delivered models (3 VAEs, the fp8 text encoder, the text projection and 2 LoRAs) run end to end on MI308X / ROCm 7.2 with no dtype, operator or kernel errors. | Measured · [R6](https://vuke.top/en/evidence/R6) [R7](https://vuke.top/en/evidence/R7) |
| Arrays to inference nodes | 1 : 8 | production sizing basis | One FX100 serves 8 inference nodes; 100 GbE sustaining KV read-back for 8 nodes is a measured anchor. On the joint-test basis, storage spend stays within roughly 10% of the GPU budget. | Estimated · [R1](https://vuke.top/en/evidence/R1) [R2](https://vuke.top/en/evidence/R2) [R3](https://vuke.top/en/evidence/R3) |

## 480B · TP8 raw readings

| Concurrency | FX100 TTFT p50 (s) | Local NVMe TTFT p50 (s) | Full recompute TTFT p50 (s) | FX100 tok/s | Local NVMe tok/s | Full recompute tok/s |
| --- | --- | --- | --- | --- | --- | --- |
| 8 | 7.53 | 10.17 | — | 56.6 | 43.9 | — |
| 16 | 11.85 | 17.31 | 149.48 | 74.9 | 53.6 | 4.1 |
| 32 | 26.35 | 35.73 | — | 71.6 | 53.9 | — |

Qwen3-Coder-480B-FP8 · TP=8 · 8 × MI308X · long-context cold-recovery load. TTFT is p50 in seconds; throughput in tok/s; full recompute was measured at concurrency 16 only. Source: [R2](https://vuke.top/en/evidence/R2)

## Test platform (R1, R2, R3, R4)

- Accelerators: 8 × AMD Instinct MI308X (192 GB HBM each, gfx942)
- Host: 2 × AMD EPYC 9654 (384 threads), about 1.5 TB RAM
- Software stack: ROCm 7.2 · vLLM 0.20.1+rocm721 · LMCache (built from upstream mainline, 2026-06-29)
- Device under test: FX100 all-flash NVMe-oF array: 4-drive RAID0 (14 TB, XFS), RoCEv2, one 100 GbE port
- Baselines: A single local NVMe drive (PCIe Gen4, 2 TB); no external storage, full recompute
- 480B model: Qwen3-Coder-480B-FP8 (MoE, about 450 GB of weights)

## Four kinds of data

- **Measured**: From a signed or official test report, with a report ID.
- **Vendor**: Nominal product specifications and list reference prices.
- **Public**: From publicly available third-party material, with a link to the source.
- **Estimated**: Derived from measured anchors, with a costing model reproducible in Python.

## Request the original reports

Tell us which report IDs you need and what for, and we will reply with how to get them. The R8 code export and raw data are provided on request in the same way.

- Company: Xinyuanyi (Tianjin) Semiconductor Equipment Co., Ltd.
- Contact: Wang Limin
- Phone: +86 131 6181 8898
- Email: mingxin@agentmail.to (The inbox is monitored and triaged by AI around the clock)

A note on names: in earlier test-report filenames FX100 appears as AISSD5000, and before that as WS5000 / GP5000; some reports are signed “Mingxin”. They all refer to the same product. This site uses the FX naming throughout (FX100 / FX200 / FX300 / FX400), while report entries keep their original titles so you can cross-check.